How can I reboot a server with systemctl if systemctl reboot fails?

Question

I tried rebooting my CentOS 7 server but it gives ridiculous error messages.

As root (of course):

# systemctl reboot
Authorization not available. Check if polkit service is running or see debug message for more information.
Failed to start reboot.target: Connection timed out
See system logs and 'systemctl status reboot.target' for details.
Exit 1

Does polkit need to check whether root has the right to reboot the machine??? If so, why?

# systemctl status reboot.target
● reboot.target - Reboot
   Loaded: loaded (/usr/lib/systemd/system/reboot.target; disabled; vendor preset: disabled)
   Active: inactive (dead)
     Docs: man:systemd.special(7)
Exit 3

Do I need to enable the reboot target? Why would this be disabled by default?

Perhaps this will work?

# systemctl start reboot.target
Authorization not available. Check if polkit service is running or see debug message for more information.
Failed to start reboot.target: Connection timed out
See system logs and 'systemctl status reboot.target' for details.
Exit 1

OK, force it, then:

# systemctl --force reboot
Authorization not available. Check if polkit service is running or see debug message for more information.
Failed to execute operation: Connection timed out
Exit 1

And the server is still up.

Answer 1

If Zach Sanchez' answer doesn't work (I got Failed to start reboot.target: Connection timed out for my systemctl --force reboot in a strange situation under CentOS 7) making the kernel basically crash reboot can be done over SSH as root like this:

# echo s > /proc/sysrq-trigger
# echo u > /proc/sysrq-trigger
# echo b > /proc/sysrq-trigger

These three commands respectively sync all mounted filesystems, remount all mounted filesystems in read-only mode, and then immediately reboot the machine.

After the last command, no response is expected as the kernel immediately reboots the machine. More details here and on Wikipedia.

As @LunarShaddow and others pointed out, a second s before b is not needed.

Answer 2

As weird as it may seem, trying running

systemctl --force --force reboot

It has popped up in a couple of searches I made. It may be related to issues with a DBus service restarting.

Can't reboot. Slow and timing out. Failed to start reboot.target: Connection timed out

Answer 3

Since I cannot make comments in Paul's answer, I write my comments here.

From the kernel document:

reboot(b) is good when you’re unable to shut down. But you should also sync(s) and umount(u) first. reboot(b) is good when you’re unable to shut down. But you should also sync(s) and umount(u) first.

sync(s) is great when your system is locked up, it allows you to sync your disks and will certainly lessen the chance of data loss and fscking. Note that the sync hasn’t taken place until you see the “OK” and “Done” appear on the screen. (If the kernel is really in strife, you may not ever get the OK or Done message...)

umount(u) is basically useful in the same ways as sync(s). I generally sync(s), umount(u), then reboot(b) when my system locks. It’s saved me many a fsck. Again, the unmount (remount read-only) hasn’t taken place until you see the “OK” and “Done” message appear on the screen.

So FMHO, Paul's answer can be simplified to 's' 'u' 'b', and you might want to wait for a while after each input.